In this project we address computer vision and deep learning for person tracking and profiling. Although we have a real application in mind as a reference for our research - the analysis of customer’s behavior inside a shopping Mall to provide personalized suggestions at digital
signage terminals - our primary goal is to investigate key scientific computer vision issues that are central to such and similar applications and develop innovative solutions with respect to the state of the art.
We will focus on the following research topics:
- Re-identification of individuals: creating anonymized identities in open-world settings to allow continous re-identification of people across different locations
Behavior analysis: understanding interests and actions of an individual
either alone or in a group
- Extraction of personal stable traits: learn hidden information such as social class, personality traits etc. To this regard we will analyze behavior and clothing to gather such profiling information
- Extraction of personal temporary feelings: understand emotional status and attention from face analysis.
To foster the usage in real contexts, we will design privacy-respectful solutions.
We plan to verify the full application in a real event at VERONAFIERE, that committed to host.